Publication: A random forest application to contact-state classification for robot programming by human demonstration
Loading...
Identifiers
Publication date
2016-03
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Wiley
Impact
Export
Abstract
This paper addresses the non-parametric estimation of the stochastic process related to the classification problem that arises in robot programming by demonstration of compliant motion tasks. Robot programming by demonstration is a robot programming paradigm in which a human operator demonstrates the task to be performed by the robot. In such demonstration, several observable variables, such as velocities and forces can be modeled, non-parametrically, in order to classify the current state of a contact between an object manipulated by the robot and the environment in which it operates. Essential actions in compliant motion tasks are the contacts that take place, and therefore, it is important to understand the sequence of contact states made during a demonstration, called contact classification. We propose a contact classification algorithm based on the random forest algorithm. The main advantage of this approach is that it does not depend on the geometric model of the objects involved in the demonstration. Moreover, it does not rely on the kinestatic model of the contact interactions. The comparison with state-of-the-art contact classifiers shows that random forest classifier is more accurate.
Description
Keywords
Multi-class contact classification, Sensor force analysis, Supervised learning
Bibliographic citation
Cabras, S., Castellanos, M. E. & Staffetti, E. (2016). A random forest application to contact-state classification for robot programming by human demonstration. Applied Stochastic Models in Business and Industry, 32(2), pp. 209–227.